|
Title:
|
EFFECTIVE RESOURCES ALLOCATION IN A P2P OVERLAY TO EXECUTE GRID WORKLOADS |
|
Author(s):
|
Rocco Aversa, Luigi Buonanno, Beniamino Di Martino, Salvatore Venticinque |
|
ISBN:
|
978-972-8939-30-4 |
|
Editors:
|
Hans Weghorn, Pedro Isaías and Radu Vasiu |
|
Year:
|
2010 |
|
Edition:
|
Single |
|
Keywords:
|
P2P, power, sharing, resources, balancing. |
|
Type:
|
Full Paper |
|
First Page:
|
69 |
|
Last Page:
|
76 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
Peer to peer systems (P2P) allow users to be connected in order to share resources and information. They have been exploited in different application contexts: distributed computing, contents and files sharing, collaborative systems. Nevertheless many proposals have been conceived to exploit P2P for sharing computational power, however each known successful solution has been designed to solve a specific problem. Many issues arise if one aims at approaching the design of a general P2P platform to support parallel/distributed computing according a Grid-like philosophy. In this paper we propose a completely distributed approach. Upon an overlay that is similar to the most popular Kademlia based P2P file sharing systems, we aim at supporting decentralized sharing of computational resources for transparent and remote execution of users applications. We focus on effective gathering of resources and on workload management of submitted tasks. Simulation results of different task allocation strategies are presented. |
|
|
|
|
|
|
